Butler University Butler University

Digital Commons @ Butler University Digital Commons @ Butler University

Scholarship and Professional Work - LAS College of Liberal Arts & Sciences

2003

Seasonal Body Temperature Fluctuations and Energetic Seasonal Body Temperature Fluctuations and Energetic

Strategies in Free-Ranging Eastern Woodchucks Strategies in Free-Ranging Eastern Woodchucks

Carmen M. Salsbury Butler University, csalsbur@butler.edu

Follow this and additional works at: https://digitalcommons.butler.edu/facsch_papers

Part of the Animal Sciences Commons, and the Biology Commons

Recommended Citation Recommended Citation Carmen M. Salsbury. "Seasonal Body Temperature Fluctuations and Energetic Strategies in Free-Ranging Eastern Woodchucks" Journal of Mammalogy 84.1 (2003): 299-310.

This Article is brought to you for free and open access by the College of Liberal Arts & Sciences at Digital Commons @ Butler University. It has been accepted for inclusion in Scholarship and Professional Work - LAS by an authorized administrator of Digital Commons @ Butler University. For more information, please contact digitalscholarship@butler.edu.

Permission to post this publication in our archive was granted by the copyright holders, Allen Press and the American Society of Mammalogists. This copy should be used for educational and research purposes only. The original publication appeared in: Carmen M. Salsbury. "Seasonal Body Temperature Fluctuations and

Energetic Strategies in Free-Ranging Eastern Woodchucks" Journal of Mammalogy 84.1 (2003): 299-310.

DOI: N/A

Journal of Mammalogy, 84(1):299-310,2003

SEASONAL BODY TEMPERATURE FLUCTUATIONS AND ENERGETIC STRATEGIES IN FREE-RANGING EASTERN

WOODCHUCKS (MARMOTA MONAX)

STAM. M. ZERVANOS* AND CARMEN M. SALSBURY

Department of Biology, Pennsylvania State University, P.O. Box 7009, Reading, PA 19610-6009, USA (SMZ)

Department of Biology, Albright College, P.O. Box 15234, Reading, PA 19612, USA (CMS) Present address of CMS: Department of Biology, Butler University, 4600 Sunset Avenue,

Indianapolis, IN 46208, USA

During a 2-year period, radiotelemetry was used to continuously monitor body temperature (Tb) of free-ranging woodchucks (Marmota monax) in southeastern Pennsylvania. Hiber­nation was preceded by daily Tb fluctuations ("test drops") of 2-4°C. During hibernation, woodchucks exhibited the characteristic pattern of torpor bouts. Time of arousals occurred randomly, but onset of torpor occurred predominantly between 1800 and 0000 h. Males had shorter hibernation periods (mean of 104.8 days) than did females (121.8 days). Males had shorter torpor bouts, but euthermic bouts were the same length as in females. Males also maintained higher T b during torpor. Overall, the cost of hibernation was greater for males than for females: males spent 38% more energy than did females. The primary energetic expense for both sexes was the periodic maintenance of euthermy throughout hibernation, which accounted for 75.2% of the energy budget for males and 66.8% for females. Compared with the 1999-2000 hibernation seasons, woodchucks during the 1998-1999 season had longer euthermic bouts, fewer torpor bouts (11.8 compared with 13.1), and spent less time in torpor (68% compared with 75%). These differences conserved more energy during the 1999-2000 hibernation season and may have been the result of severe drought conditions during summer 1999. After emergence from hibernation, woodchucks generally maintained a constant state of euthermy throughout the active season, with Tb fluctuating daily by I-2°C. However, during the summer drought of 1999, daily Tb fluc­tuated 8-17°C in 5 of 8 woodchucks, presumably to conserve energy and water.

Key words: body temperature, energetics, estivation, hibernation, Marmota monax, torpor, woodchucks

Hibernation is prominent among ground­dwelling squirrels of the family Sciuridae and is characterized by an extreme reduc­tion in metabolic function and body tem­perature (Tb) that may extend over repeated bouts, each lasting several days (Lyman et al. 1982). Because of substantial energy savings from depressed metabolic function, hibernation is widely used as a mechanism to survive harsh conditions such as low am­bient temperatures (Ta) and food scarcity.

* Correspondent: smzl@psu.edu

299

The hibernation patterns of ground­dwelling squirrels have been studied exten­sively in the laboratory (French 1982; Geis­er and Kenagy 1988; Harlow and Menkens 1986; Pivorun 1976). Fluctuations in Tb and arousal state have been documented for many species in captivity including species of Marmota (Armitage et al. 2000; French 1990; Heldmaier et al. 1993; Lyman 1958). Depletion of fat stores and changes in body mass throughout hibernation have also been examined (Armitage et al. 2000; Davis

300 JOURNAL OF MAMMALOGY Vol. 84, No.1

1967; Florant 1998; Snyder et al. 1961). Further, energy expenditure as estimated by O2 consumption has been monitored in a large variety of species in the laboratory setting (Geiser 1988).

Field studies of hibernation have gener­ally involved an assessment of immergence and emergence dates based on trapping in­formation and field observations (Davis 1976; Ferron 1996; Michener 1977). Loss of body mass has been estimated for free­ranging hibernators by comparing body mass of animals captured late in the active season with body mass of animals shortly after emergence from hibernation (Arnold 1993; Barash 1989; Buck and Barnes 1999a, 1999b; Ferron 1996). The energetic cost of hibernation has been inferred from estimates of daily mass loss during hiber­nation. However, automated radiotelemetry monitoring of Tb of free-ranging hiberna­tors (Zervanos and Salsbury 2001) may provide further insight into the patterns of hibernation in the field and may validate the relevance of the wealth of laboratory infor­mation on hibernation patterns that have been collected for a number of species.

The continual observation of Tb of hi­bernators may also afford investigators the opportunity to estimate energy expenditure of free-ranging hibernators. For example, one method for monitoring CO2 production in animals in their natural environments is the use of doubly labeled water (Nagy 1989). The utility of this method for mea­suring the energetic costs of the different phases of hibernation-arousal, euthermy, and torpor-is limited because the resolu­tion of this method is generally restricted to a period of several days and requires 2 blood samples from the focal animal. How­ever, by monitoring Tb of free-ranging hi­bernators, energy expenditures in the field can be extrapolated from known relation­ships between metabolism and T b generated under laboratory conditions. This technique has been used successfully to estimate en­ergetic costs of hibernation for free-ranging Richardson's ground squirrels, Spermophi-

Ius richardsonii (Wang 1972, 1973, 1978); however, surprisingly, few other field stud­ies have been conducted that provide the detail necessary to estimate energy expen­diture.

We monitored T b of eastern woodchucks (M. monax) throughout hibernation and during the active season in an attempt to characterize and better understand T b pat­terns under natural conditions. Additionally, we hoped to verify data generated from lab­oratory studies of hibernation patterns of woodchucks. Finally, we wanted to gener­ate an estimate of energy expenditure for hibernation of free-ranging woodchucks by extrapolating laboratory data to field con­ditions.

MATERIALS AND METHODS

Trapping and field observations.-The study was conducted from 29 August 1998 to 19 Sep­tember 2000 on the Peiffer Farm of the Penn­sylvania State University's Berks Campus locat­ed in southeastern Pennsylvania (400 22'N, 75°22'W). About 35 ha of the 44-ha farm were planted with corn (maize) each summer. Three woodlots (mixed hardwoods), encompassing an area of approximately 4 ha, are contained within the farm boundaries. The remainder of the farm consists of open fields in various stages of suc­cession, experimental study plots, and farm and maintenance buildings. An estimated population of 46 woodchucks (M. monax) inhabits the farm, primarily along the ecotones of the woodlots and fields.

Woodchucks were trapped 5 days/week from emergence in March until immergence in Oc­tober of each year. Trapped animals were sexed, weighed, tagged, and then released. Times of emergence during the hibernation period were determined by placing straw in all the entrances of inhabited burrows. The entrances were ex­amined 3 times/week, and any disturbance of the straw was assumed to be an emergence.

Radiotelemetry system.-Radiotelemetry was used to monitor hourly Tb of free-ranging adult woodchucks (Zervanos and Salsbury 2001). At any given time during the study, 6-14 wood­chucks were simultaneously monitored using surgically implanted intraperitoneal temperature transmitters (model IMP/3001L, Telonics, Mesa,

February 2003 ZERVANOS AND SALSBURY-HIBERNATION OF WOODCHUCKS 301

Arizona). Temperature transmitters (model MOD-20S, Telonics) were also placed 2-3 m into 4 woodchuck burrows to monitor burrow temperatures. To determine the placement of the transmitters in the burrows, a 6-m graduated string (marked at I-m intervals) was tied to the transmitter that was then taped to the tail of a captured woodchuck. The animal was released at the entrance of its burrow, and when it stopped moving, the distance entered was mea­sured and the string was pulled to release the transmitter. Another transmitter was placed in an outdoor weather station to monitor Ta. All trans­mitters provided a 0.2°C resolution through the entire period of data collection. Telemetry data were automatically recorded with a data acqui­sition system (Zervanos and Salsbury 2001). In addition, each animal was manually tracked and located on a weekly basis with portable telem­etry equipment.

Weather data (Ta' wind velocity and direction, humidity, and rainfall) were also collected au­tomatically each hour the year round by an out­door weather station and were stored in a com­puter file.

Data analysis.-We examined torpor, arousal, and euthermic bouts for each individual through­out hibernation. We defined euthermy as Tb > 30°C and torpor as T b < 30°e. We defined deep torpor as the period of low stable Tb < 20°e. We defined hibernation season as starting with the I st deep torpor in autumn and ending with the last arousal in spring and active season as the rest of the year. During the hibernation sea­son, time of day for each arousal was measured as the point at which the Tb started to rise above the Tb of stable deep torpor, and beginning of decline into torpor was measured as the point at which Tb began to decline below stable euther­mic Tb. We used these data to test randomness of starting time for both arousal from and de­cline into torpor. We divided the day into four 6-h periods of observations starting at midnight (0000 h). A chi-square analysis was performed assuming an expected equal distribution of start­ing times within each 6-h time period.

We calculated arousal and cooling rates eCI min and °C min-I kg-I) for each torpor arousal sequence for which we had interpretable, contin­uous data, and values were averaged for each animal. Arousal rate was calculated from time in hours for Tb to rise from stable deep torpor to 30°C, and cooling rate was calculated from

time in hours for Tb to fall from 30°C to a stable deep torpor temperature. Because the absolute decline of Tb was not the same for all individ­uals, we also calculated mean cooling rate from 30°C to 20°C for each individual. We examined relationships of arousal and cooling rates to bur­row temperatures by calculating Spearman cor­relation coefficients. Arousal and cooling rates, as well as total time to arouse and cool, were compared between sexes using Student's t-tests.

The hibernation season of each animal for which we collected interpretable data was divid­ed into 3 phases of equal duration--early, mid, and late. Mean Tb during torpor, mean Tb during euthermy, mean torpor length (h), and mean euthermy length (h) were calculated for each an­imal during each phase. These 4 variables were analyzed in separate repeated-measures 3-way analyses of variance (ANOV A), where sex, year, and hibernation phase (repeated measure nested within sex) were designated as main effects in a general linear model. Significance was assigned based on a probability of a < 0.05. Hibernation phase, when not involved in significant interac­tions, was examined in post hoc tests using the Tukey-Kramer multiple comparisons method. All repeated-measures ANOV A and post hoc tests were performed using Minitab statistical software (Release 13 for Windows, Minitab In­corporated, State College, Pennsylvania).

Body mass data were averaged by month for each sex. We compared mean monthly body mass between sexes by using Student's t-statis­tics. Because animals did not emerge from their burrows, it was not possible to weigh them dur­ing hibernation; therefore, we estimated rate of daily mass loss through hibernation by subtract­ing posthibernation body mass from prehiber­nation body mass and dividing by length of hi­bernation. Because animals were difficult to trap just before and after hibernation, daily mass loss estimates were based on mean body masses col­lected for the population in October and March and mean length of hibernation for each sex. We also calculated mass-specific daily mass loss (mg g [immergence mass]-t day-I) during hi­bernation for each sex.

RESULTS

Seasonal climate conditions.-Mean bur­row temperatures ranged from summer highs of 20.3°C to winter lows of 6.3°C. These seasonal changes were gradual and

302 JOURNAL OF MAMMALOGY Vol. 84, No.1

reflect the relative protection of burrows from seasonal extremes. Mean burrow tem­perature during June-August was higher in 1999 (16.9 SE :::+::: 0.4°C) than in 2000 (15.3 :::+::: 0.2°C: t = 1.94, d.f = 172, P < 0.05). Mean burrow temperatures during the hi­bernation season in 1998-1999 (7.3 :::+:::

0.2°C) and 1999-2000 (7. 7 :::+::: 0.2°C) were not different (t = 1.13, d.f = 196, P > 0.10). Daily burrow temperature fluctuated about 1.8°C during both summers and about 1.0°C daily during both hibernation sea­sons.

Except for 2 periods, seasonal ambient conditions were similar throughout the study. The 1st exception was for Decem­ber-February 1999-2000, when mean Ta was significantly lower (0.8°C) than for the same period in 1998-1999 (2.9°C). For the 2nd exception, June-August 1999, rainfall was at a record low (10.3 cm) compared with the same 3-month period in 2000 (30.1 cm), vegetation was dry and unproductive, and freestanding water was only available during limited rainfalls or from a creek along the eastern boarder of the study area. Activities around the farm buildings also offered some drinking water. However, only 3 of the study animals were within 45 m of any of these water sources; the rest were at least 250 m away from a permanent water source. The drought ended in September when 20.8 cm of rain fell over a 26-day period.

Seasonal Tb patterns.-Hibernation typ­ically started with daily fluctuations of Tb ("test drops" -Pivorun 1976) several days before the 1st torpor bout (Fig. la). These fluctuations ranged between 2°C and 4°C, with a steady decline in the euthermic range. The majority (85%) of the daily high Tbs were reached between 1400 and 1700 h. The majority (85%) of daily low Tbs were reached between 0400 and 0600 h. This general pattern was observed for all animals during both autumn seasons. A sig­nificant number (69%) of declines into the 1 st torpor bout occurred between 1800 and 0600 h (X2 = 3.85, d.! = 1, P < 0.05).

6 ~ 30 ~ ::I 16 20

~ E 10

~ Ambient temperature

29 Oct 31 Oct 2 Nov 4 Nov 6 Nov 8 Nov 10 Nov 12 Nov

6 ~ 30 ~ ::I 16 20

~ E 10

~

12 0cI 16 Oct 20 OCI 24 Oct 28 Oct t Nov 5 Nov 9 Nov

6' ~ 30

~ ::I "§ 20 (]) Co E 10

~

23 Feb 25 Feb 27 Feb 29 Feb 2 Mar 4 Mar 6 Mar 8 Mar

Date

FIG. I.-a) Body temperature fluctuations of female Marmota monax (number 1) before ini­tial torpor bout of 1999-2000 hibernation sea­son. b) Torpor bouts of female M. monax (num­ber 2) at beginning of the 1999-2000 hiberna­tion season. c) Body temperature fluctuations of female M. monax (number 3) before last arousal of 1999-2000 hibernation season. Burrow (av­eraged from 4 burrow locations) and ambient temperatures are also presented.

During hibernation, woodchucks exhib­ited the characteristic pattern of torpor bouts (Fig. Ib). During each successive bout, deep torpor T b declined with declining burrow temperature, reaching winter lows of 6.5-8.4°C during 1998-1999 and 6.8-8.8°C during 1999-2000. Animals aroused randomly throughout the day with no dis­cernable pattern. However, time of entry into torpor was not randomly distributed;

February 2003 ZERVANOS AND SALSBURY-HIBERNATION OF WOODCHUCKS 303

41 % occurred between 1800 and 0000 h (X2

= 8.66, d.! = 3, P < 0.05). Final emergence from hibernation was

usually preceded by a relatively long euth­ermic period of 3-4 days in 15 of 18 ani­mals observed. The 3 exceptions exhibited shorter euthermic periods of 1-2 days. Nine of the 15 animals with long euthermic pe­riods were characterized by daily fluctua­tions in Tb (Fig. lc). The brief but signifi­cant declines reached a low Tb (23-28°C) between 2100 and 0300 h. A significant number (67%) of arousals from the final torpor bout occurred during the day (X 2 = 5.44, d.! = 1, P < 0.02).

After emergence, most animals main­tained a constant state of euthermy through­out the active season with T b fluctuating daily by I-2°C. Excluding animals with daily Tb fluctuations >2°C, April 2000 mean daily T b was significantly lower (37.1°C) than in July 2000 (37.6°C; t = 2.93, dj. = 60, P < 0.01). However, some adults of both sexes exhibited daily Tb fluc­tuations in excess of 10°C. Between 10 March 1999 and 6 May 1999, 3 of 8 ani­mals had daily Tb fluctuations of 1O-14°C. Between 2 April 2000 and 4 June 2000, 2 of 10 animals had daily Tb fluctuations of about 10°C. We were unable to find envi­ronmental correlates to these fluctuations. However, during summer of 1999, daily Tb fluctuated 8-l7°C in 5 of 8 woodchucks (Fig. 2a). These fluctuations occurred be­tween 2 August and 17 September and were associated with severe drought and high temperature conditions during this period. One of the 5 animals died during this pe­riod. Starting in mid-August, significant rain (> 1 cm in 24 h) fell, and 2 animals returned to stable euthermia within 2 days, whereas 2 animals continued to fluctuate Tb by 6-l2°C until mid-September. Of the 3 woodchucks with stable euthermia during August 1999, 2 lived in burrows within 40 m of the creek, and 1 inhabited a watered vegetable garden. During the same period in 2000, rainfall and Tas were normal, and

40

IT ~30

~ :J ai 20 -~

(I) Co E 10

~

2Aug 4Aug 6Aug 8Aug 10Aug 12Aug 14Aug 16Aug

IT ~30 (I) ~

:J ai 20 ~

(I)

~ 10

~ 2Aug 4Aug 6Aug 8Aug lOAug 12Aug 14Aug 16Aug

Date FIG. 2.-a) Body temperature fluctuations of

female Marmota monax (number 4) during sum­mer drought of 1999. Arrow indicates rainfall of 3.6 cm in a 28-h period. This was the 1st sig­nificant rainfall (> 1 cm in 24 h) in 56 days. b) Body temperature fluctuations of female M. monax (number 5) during summer of 2000 with­out drought. Burrow (averaged from 4 burrow locations) and ambient temperatures are also presented. (Temperature labels as for Fig. la.)

none of the woodchucks exhibited daily Tb fluctuations >2°C (Fig. 2b).

Analysis of hibernation.-Male and fe­male woodchucks living in our study area had similar body masses in March after hi­bernation and in October just before hiber­nation (Table 1). During the active season, males were significantly heavier than fe­males only in June (2-tailed t-test: t = 3.18, dj. = 25, P = 0.009). The population es­timate of daily mass loss during hibernation was 2.11 mg g-I day-I for males and 1.78 mg g-l day-I for females.

Males had shorter hibernation periods (Table 1). Both males and females showed similar cooling rates, but length of the cool­ing phase was significantly shorter for males. Males and females did not differ in

304 JOURNAL OF MAMMALOGY Vol. 84, No.1

TABLE l.-Comparison of hibernating male and female woodchucks. Dates are reported in Julian calendar days. Asterisk denotes significant differences (P ::s 0.05, I-test).

Females Males

n X SE n X SE P

Body mass (kg)

March 11 2.79 0.179 3 2.84 0.549 0.456 October 10 3.57 0.256 5 3.64 0.172 0.425

Date of 1 st torpor 14 305.7 1.94 5 314.6 13.30 0.296 Date of last arousal 14 62.5 1.95 5 58.4 4.41 0.338 Number of torpor bouts 14 12.7 0.40 5 12.0 1.14 0.460 Length of hibernation (days) 14 121.8 1.83 5 104.8 8.82 0.010* Arousal length (h) 9 4.3 0.23 3 4.3 0.67 0.944 Length of cooling phase (h) 9 66.3 3.70 3 52.1 3.20 0.023* Arousal rate (OC/min) 9 Arousal rate (OC min- 1 kg-I) 9 Cooling rate (OC/min) 9 Cooling rate between 30°C and 20°C (OC/min) 9

number of torpor bouts during hibernation, length of each arousal phase, or rate of arousal (Table 1). Neither cooling nor arousal rate was significantly correlated with burrow temperature (rs = -0.172, df = 31, P > 0.05 and rs = -0.280, d.! = 40, P > 0.05, respectively).

0.100 0.0048 3 0.103 0.0110 0.851 0.028 0.0027 8 0.031 0.0061 0.707 0.0067 0.0003 3 0.0085 0.0015 0.371 0.0241 0.0015 3 0.0247 0.0081 0.948

Repeated-measures ANOV A indicate that males engaged in significantly shorter torpor bouts throughout the hibernation pe­riod and maintained a higher Tb during tor­por than did females (Table 2). There were no significant differences between sexes with regard to euthermy. Torpor length and

TABLE 2.-Effects of sex, year, and hibernation phase on hibernation characteristics of woodchucks. Means and SE are presented for the main effects. P ::s 0.05 was considered significant.

Length of Tb during

Torpor bouts (h) Euthermy bouts (h) Torpor (0C) Euthermy (OC)

n X SE n X SE n X SE n X SE

Sex

Male 5 144.5 12.6 5 66.5 5.9 5 11.3 0.7 5 35.4 0.9 Female 14 170.9 6.5 14 64.6 4.3 14 10.2 0.5 14 35.7 0.2 F 7.92 0.26 5.05 0.01 P 0.007 0.614 0.030 0.923a

Year

1998-1999 9 166.8 8.0 9 76.5 6.0 9 10.5 0.6 9 35.6 0.3 1999-2000 10 161.5 8.9 10 54.8 2.7 10 10.5 0.6 10 35.7 0.5 F 0.09 8.80 0.00 0.64 p 0.764b 0.005 0.973' 0.428a

Hibernation phase

Early 19 127.6 7.6 19 56.9 3.9 19 14.2 0.4 19 35.2 0.6 Mid 19 192.9 7.6 19 56.0 7.4 19 8.7 0.3 19 35.7 0.5 Late 19 171.4 9.8 19 63.5 5.9 19 8.6 0.5 19 36.0 0.3 F 10.07 3.96 43.56 0.64 p <O.OOl b 0.008 <0.001' 0.637

, F = 4.08, P = 0.049 for significant interactions. b F = 3.05, P = 0.026 for significant interactions. C F = 3.52, P = 0.014 for significant interactions.

February 2003 ZERVANOS AND SALSBURY-HIBERNATION OF WOODCHUCKS 305

T b during torpor and euthermy did not vary significantly between the 2 hibernation sea­sons. Euthermy bouts were significantly longer during the 1998-1999 season than during the 1999-2000 season.

Length of torpor bout, length of euther­my bout, as well as T b during torpor varied significantly across the 3 phases of hiber­nation (Table 2). Post hoc multiple com­parisons of length of euthermy indicate that there is a trend toward longer euthermic bouts during the late phase of hibernation than during the early phase of hibernation, but the difference was not significant (t =

2.88, P = 0.0628). Interpretation of the var­iation in length of torpor bouts and Tb dur­ing torpor across the hibernation phases is difficult due to significant interactions be­tween hibernation phase and year for both dependent variables. Post hoc examination of interactions suggests there is a trend, es­pecially among females, toward shorter tor­por bouts and higher Tb early in hibernation than in mid and late hibernation. These trends, however, were not consistent be­tween sexes or across years.

Several other differences were observed between years. Average date of immer­gence was 7 November during 1998-1999 compared with 1 November during 1999-2000 (t = 2.26, d.f. = 17, P < 0.05). Av­erage date of emergence was 6 March dur­ing 1998-1999 and 27 February for 1999-2000 (t = 2.42, dj = 17, P < 0.05). The average number of torpor bouts was signif­icantly different, 11.8 for 1998-1999 com­pared with 13.1 for 1999-2000 (t = 2.91, dj = 17, P < 0.01). Percentage of time spent in torpor was significantly less in 1998-1999 than in 1999-2000, 68% com­pared with 75% (t = 2.98, dj = 17, P < 0.01). However, average length of hiber­nation was not significantly different be­tween hibernation seasons, 119 days for 1998-1999 compared with 115 days for 1999-2000 (t = 1.19, d.! = 17, P = 0.19).

DISCUSSION

Active season strategies.-During the ac­tive season, woodchucks displayed a typical

euthermic pattern of small daily Tb fluctu­ations (Fig. 2b). Hayes (1976) observed similar patterns in woodchucks from Ar­kansas. He also found, as we have, that mean daily Tb is significantly lower (by 0.5°C) in spring than in midsummer. Hayes concluded that higher T b during summer re­duced the thermal radiation gradient and fa­cilitated passive heat loss. Lower Tb in spring would minimize heat loss and save thermoregulatory energy during colder spring days. Under environmental stress, woodchucks increased daily Tb fluctuations such that nightly Tb approached torpor lev­els. Spring Tb fluctuations occur during pe­riods when food availability is low and en­ergy cost for reproduction is high (Fall 1971; Snyder et al. 1961). Similar fluctua­tions in Tb were observed in August 1999 during a prolonged drought (Fig. 2a). This type of adaptive response has not, to our knowledge, previously been observed under field conditions in marmots (Davis 1976). However, Davis (1967, 1976) found that captive woodchucks can enter torpor at any time of the year, provided they are deprived of food and temperatures are low (6°C). He found that woodchucks kept at 20°C and deprived of food do not become torpid but become "lethargic" and lower Tb to 33°C. During the drought period in our study, woodchucks were rarely observed above­ground and, when they were, it was only during the cooler early morning or early evening hours. The lower Tb and restricted aboveground activity would conserve met­abolic energy and reduce thermoregulatory water loss during periods of drought.

Dynamics of hibernation.-Test drops, similar to those in our study, have been ob­served during initial torpor in the eastern chipmunks (Pivorun 1976), alpine marmots (Arnold 1988), and woodchucks (Lyman 1958). Yet their function is unknown. Be­cause these test drops precede only the 1st torpor bout, they may reflect the daily re­setting of the hypothalamic thermostat and function to acclimate cellular functions to the onset of low Tb (Hammel 1967; Luecke

306 JOURNAL OF MAMMALOGY Vol. 84, No.1

and South 1972). This concept is supported by observations made by Pivorun (1976) who found that the number of test drops for chipmunks increased with decreasing Ta. Thus, the lower the reset requirement, the more test drops that would be needed.

After the drought summer of 1999, woodchucks aroused more often but stayed euthermic less time per bout. As a result, length of hibernation remained the same be­tween the 2 hibernating seasons, but per­centage of time spent in torpor was higher for 1999-2000. The severe drought condi­tions during summer of 1999 could have influenced this pattern. Early immergence and emergence during the 1999-2000 sea­son, together with higher percentage of time in torpor, would be consistent with the need to conserve more energy after a summer of low food resources. Early immergence may allow animals to conserve energy under poor forage conditions, whereas early emer­gence may allow animals to assess the en­vironment sooner for potential resources. Mean daily maximum Ta between mid-Feb­ruary and mid-March was 6.5°C in 1999 compared with 13SC in 2000. However, it is not possible to assess whether environ­mental conditions, circannual rhythms, or nutrient needs were responsible for early emergence (Davis 1976, 1977; Pengelley 1967). Finally, percentage of time spent in torpor might be more important in total en­ergy conservation than in the energy loss from more arousals (French 1992).

Our inability to demonstrate a relation­ship between burrow temperatures and arousal rates is consistent with the findings of Wang and Hudson (1971) for eastern chipmunks. Lyman (1958) and Wang (1978) suggest that the mechanism control­ling differential warming of body parts dur­ing arousal functions irrespective of Ta. However, temperature acclimation may also playa role. Squirrels kept at 15°C arouse more slowly than those kept at 5°C (French 1982). Similar relationships have been found for other mammals (Geiser and Bau-

dinette 1987; Zervanos and Henshaw 1970).

Cooling rates were not correlated with burrow temperatures for woodchucks in our study. Lyman (1958), Pivorun (1976), and Wang (1978) also found no relationships between cooling rates and ambient (burrow) temperatures. Pivorun (1976) observed that the cooling rate for a dead chipmunk is al­ways significantly faster than that for the same animal alive and entering torpor. He attributed this to the existence of a control on cooling to prevent cellular damage from a too-rapid decline in Tb. Lyman (1958) at­tributed this phenomenon to vascular con­trols causing differential cooling of body core and periphery. If cooling is not pas­sive, it would explain the lack of correlation between cooling rates and burrow temper­atures.

Biological rhythms.-During the active season, high Tb during afternoon hours for woodchucks is consistent with diurnal ac­tivity patterns and is in agreement with fluc­tuations found for other ground squirrels (Michener 1992; Wang 1972). During hi­bernation, woodchucks arouse from torpor randomly but enter torpor primarily be­tween 1800 and 2300 h. Canguilhem et al. (1994) found preferential times between 0000 and 0800 h for torpor entrance in the European hamster and a random pattern for arousal times. They speculate that random­ness of arousal and rhythmicity of entrance into torpor indicate a resynchronization of the internal clock during arousal periods. They did not identify an entrainer (zeitge­ber) for resynchronization. Because bur­rows are sealed during hibernation, photo­period entrainment is not possible (Davis 1976). Although the exact location of our temperature transmitters within the burrow system is unknown, the daily 1°C fluctua­tion in burrow temperature could have act­ed as an entrainer (Pittendrigh 1960; Zer­vanos 1969). During our study, both males and females were observed to 1 st emerge from their burrows several weeks before their final arousal from hibernation. These

February 2003 ZERVANOS AND SALSBURY-HIBERNATION OF WOODCHUCKS 307

early forays occur in mid to late afternoon, and most animals initially stay within 2 m of the burrow entrance. They are unable to feed during this time due to snow cover and absence of edible vegetation. Thus, it is possible that early emergence behavior may be due, in part, to the need to resynchronize the hibernator's biological clock. The syn­chrony of final arousals from hibernation, the majority of which occur during daytime hours, indicates that a rhythmic pattern has been established by the end of hibernation. This supports the conclusion that some rhythmic behavior is maintained during hi­bernation (Canguilhem et al. 1994; Davis 1976) and may be further synchronized just before the end of hibernation.

Energetics of hibernation.-Knowing the hibernation patterns of free-ranging wood­chucks provides a rare opportunity to verify data collected for this species under captive conditions. Also, by estimating energetic cost of hibernation in the field, it may be possible to ultimately gain insight into other aspects of woodchuck life histories such as overwinter mortality rates and variable re­productive success among individuals and seasons. Although we did not examine hi­bernation patterns of captive woodchucks in this study, woodchucks from the same study population were simultaneously ex­amined in captivity by Armitage et al. (2000). These authors monitored hiberna­tion state of individuals throughout hiber­nation and measured the oxygen consump­tion of hibernating woodchucks as they passed through periods of arousal, euther­my, and torpor at 6°C. Woodchucks hiber­nating in captivity displayed patterns of arousal and torpor similar to those of free­ranging hibernators. Captive animals en­tered torpor less frequently (X = 10.2 ± 0.85 bouts-K. B. Armitage, pers. comm.) throughout hibernation than did free-rang­ing hibernators (X = 12.4 ± 0.41 bouts; Student's t-test P = 0.053). The lower num­ber of torpor bouts among captive animals may be due, in part, to the short hibernation period of some of the captive hibernators

(X = 99.6 ± 10.3 days-K. B. Armitage, pers. comm.). Others have also found that hibernation periods are typically shorter for captives than for free-ranging animals (Heldmaier et al. 1993). Armitage et al. (2000) report a mean time spent in torpid of 52.9% for captive woodchucks on a monthly basis from November through March. This value is considerably less than estimates for free-ranging hibernators (males X = 68.9% and females X =

72.8%); however, the estimate for captive animals is similar when the value is recal­culated on an individual basis and periods of pre hibernation euthermy are not included in the estimate (X = 69.3%-K. B. Armi­tage, pers. comm.). The decline in Tb to deep torpor after a euthermic period was much more rapid for animals in captivity (X = 33.4 h-Arrnitage et al. 2000) than for animals in the field (overall for males and females: X = 63.3 h). The difference in cooling rates may be explained by the fact that measurement of oxygen consump­tion in the laboratory involved an active flow of air over the hibernator (Armitage and Salsbury 1992) that may have in­creased convection and evaporative heat loss. Naturally hibernating woodchucks are known to use vegetation to plug the en­trances of their burrow system (Grizzell 1955); this behavior would presumably minimize airflow and reduce passive heat loss.

We calculated metabolic costs of hiber­nation for each field-monitored animal us­ing the following equation from French (1992): metabolic cost of hibernation =

([total hours of euthermy][ml O 2 g-l h-1Hpreimmergence body mass]) + ([total hours of torporHml O2 g-l h-1Hpreimmerg­ence body mass]) + ([total number of arousalsHml Oiarousal]). The metabolic rates for woodchucks (euthermy = 0.141 ml O2 g-l h- 1, torpor = 0.014 ml O2 g-I h- 1, arousal = 0.229 ml O2 g-I h- 1) were obtained from Armitage et al. (2000). Met­abolic cost (ml O2) of each arousal phase was estimated for each individual by mul-

308 JOURNAL OF MAMMALOGY Vol. 84, No.1

tiplying the mean arousal length (time in hours to increase T b from deep torpor to 30°C) by the weight-specific cost per hour and the preimmergence mass. We assumed that the previously mentioned similarities in hibernation patterns of captive and field an­imals justified our extrapolation from lab­oratory to field setting. The metabolic costs for field hibernators are most likely over­estimates because we were unable to weigh animals during hibernation in the field. Thus, mass lost during hibernation and any resultant decrease in total metabolism were not accounted for in the calculations. Also, faster cooling rates of woodchucks in cap­tivity than in the field and the lower T. in the laboratory study (6°C) than in the field (7-8°C) suggest that laboratory estimates may be higher than energetic costs incurred by animals under more insulated conditions in the field. Given that, one cannot correct for these factors; these metabolic cost esti­mates may be best used for relative com­parisons among individuals rather than for absolute comparisons between studies.

Males spent 38% more energy for hiber­nation than did females (males it = 736.3 liters of O2 and females it = 531.9 liters of 02)' Males also spent 56% more energy on a daily basis during hibernation than did fe­males. The primary energetic expense for both sexes resulted from periodic euthermy throughout hibernation, which accounted for 75.2% of the energy budget for males and 66.8% for females. The greater energy expenditure by males for euthermy was the result of the greater proportion of time males spent euthermic in the field than did females. Why do males not use torpor to the same extent as females during hiber­nation? Unfortunately, our results do not provide the information necessary to ad­dress this question. It has been suggested for other species of ground squirrels, how­ever, that euthermic Tb is necessary to sup­port spermatogenesis (Barnes et al. 1986), which begins before hibernation ends (Christian et al. 1972). Male woodchucks also may benefit from maintaining euther-

my during late hibernation because it may allow them periodically to leave their bur­rows and assess climatic conditions for op­timal timing of emergence. Male wood­chucks in this study and others (Davis 1977; Grizzell 1955; Snyder and Christian 1960) have been observed aboveground for short periods in January and February. The earliest possible emergence from hiberna­tion may be critical for males because re­productive success is dependent on their ability to establish territories, locate fe­males, and complete spermatogenesis be­fore females emerge. Thus, the higher cost of hibernation for males may result in sub­stantial fitness payoffs as they emerge from hibernation physiologically ready to locate females and to mate. Females apparently adopt a more energy-conservative hiberna­tion strategy consisting of longer torpor bouts and later spring emergence, preserv­ing their energy stores for the energetic de­mands of reproduction.

The variation in strategies for hibernation practiced by males and females in this study emphasizes the need to examine more care­fully the adaptational issues related to the synchrony of hibernation and the physio­logical mechanisms of torpor and arousal. Further, individual and yearly variations in T b measurements indicate that hibernation patterns may vary with climatic conditions. Thus, application of the present Tb-moni­toring technique to M. monax and other hi­bernating species in different habitats may help to further elucidate the relationships between climatic factors and hibernation patterns in the wild.

ACKNOWLEDGMENTS

We thank K. B. Armitage for his helpful re­view of our manuscript, J. Brown for her assis­tance in data collection and analysis, and A. Chapdelaine for statistical advice. We also thank a number of students from Pennsylvania State University and Albright College whose help in the field was invaluable. The study was sup­ported by Faculty Development Grants from the Pennsylvania State University and Albright Col­lege.

February 2003 ZERVANOS AND SALSBURY-HIBERNATION OF WOODCHUCKS 309

LITERATURE CITED

ARMITAGE, K. B., AND C. M. SALSBURY. 1992. Factors affecting oxygen consumption in wild-caught yel­low-bellied marmots. Comparative Biochemistry and Physiology, A. Comparative Physiology 103: 729-737.

ARMITAGE, K. B., B. C. WOODS, AND C. M. SALSBURY. 2000. Energetics of hibernation in woodchucks (Marmota monax). Pp. 73-80 in Life in the cold: Eleventh International Symposium (G. Heldmaier and M. Klingenspor, eds.). Springer-Verlag, Berlin, Germany.

ARNOLD, W. 1988. Social thermoregulation during hi­bernation in alpine marmots (Marmota marmota). Journal of Comparative Physiology, B. Biochemical, Systematic, and Environmental Physiology 158: 151-156.

ARNOLD, W. 1993. Energetics of social hibernation. Pp. 65-80 in Life in the cold: ecological, physiological, and molecular mechanisms (C. Carey, G. L. Fiorant, B. A. Wunder, and B. Horwitz, eds.). Westview Press, Boulder, Colorado.

BARASH, D. P. 1989. Marmots: social behavior and ecology. Stanford University Press, Stanford, Cali­fornia.

BARNES, B. M., M. KRETZMANN, P. LICHT, AND I. ZUCK­ER. 1986. The influence of hibernation on testis growth and spermatogenesis in the golden-mantled ground squirrel, Spermophilus lateralis. Biology of Reproduction 35:1289-1297.

BUCK, C. L., AND B. M. BARNES. 1999a. Annual cycle of body condition and hibernation in free-living Arc­tic ground squirrels. Journal of Mammalogy 80: 430-442.

BUCK, C. L., AND B. M. BARNES. 1999b. Temperatures of hibernacula and changes in body composition of arctic ground squirrels over winter. Journal of Mam­malogy 80:1264-1276.

CANGUILHEM, B., ET AL. 1994. Search for rhythmicity during hibernation in the European hamster. Journal of Comparative Physiology, B. Biochemical, Sys­tematic, and Environmental Physiology 163:690-698.

CHRISTIAN, J. J., E. STEINBERGER, AND T. D. McKINNEY. 1972. Annual cycle of spermatogenesis and testis morphology in woodchucks. Journal of Mammalogy 53:708-716.

DAVIS, D. E. 1967. The role of environmental factors in hibernation of woodchucks (Marmota monax). Ecology 48:683-689.

DAVIS, D. E. 1976. Hibernation and circannual rhythms of food consumption in marmots and ground squirrels. Quarterly Review of Biology 51: 477-514.

DAVIS, D. E. 1977. Role of ambient temperature in emergence of woodchucks (Marmota monax) from hibernation. American Midland Naturalist 97:224-229.

FALL, M. W. 1971. Seasonal variations in the food con­sumption of woodchucks (Marmota monax). Journal of Mammalogy 52:370-375.

FERRON, J. 1996. How do woodchucks (Marmota mon­ax) cope with harsh winter conditions? Journal of Mammalogy 77:412-416.

FLORANT, G. L. 1998. Lipid metabolism in hibernators: the importance of essential fatty acids. American Zoologist 38:331-340.

FRENCH, A. R. 1982. Intraspecific differences in the pattern of hibernation in the ground squirrel Sper­mophilus beldingi. Journal of Comparative Physiol­ogy, B. Biochemical, Systematic, and Environmental Physiology 148:83-91.

FRENCH, A. R. 1990. Age-class differences in the pat­tern of hibernation in yellow-bellied marmots, Mar­mota flaviventris. Oecologia 82:93-96.

FRENCH, A. R. 1992. Mammalian dormancy. Pp. 105-121 in Mammalian energetics (T. E. Tomasi and T. H. Horton, eds.). Cornell University Press, Ithaca, New York.

GEISER, F. 1988. Reduction of metabolism during hi­bernation and daily torpor in mammals and birds: temperature effect or physiological inhibition? Jour­nal of Comparative Physiology, B. Biochemical, Systematic, and Environmental Physiology 158:25-37.

GEISER, F., AND R. V. BAUDINETTE. 1987. Seasonality of torpor and thermoregulation in three dasyurid marsupials. Journal of Comparative Physiology, B. Biochemical, Systematic, and Environmental Phys­iology 157:335-344.

GEISER, F., AND G. J. KENAGY. 1988. Torpor duration in relation to temperature and metabolism in hiber­nating ground squirrels. Physiological Zoology 61: 442-449.

GRIZZELL, R. A. 1955. A study of the southern wood­chuck, Marmota monax monax. American Midland Naturalist 53:257-293.

HAMMEL, H. T. 1967. Temperature regulation and hi­bernation. Pp. 86-96 in Mammalian hibernation III (K. C. Fisher, A. R. Dawe, C. P. Lyman, E. Schon­baum, and F. E. South, eds.). Oliver and Boyd, Lon­don, United Kingdom.

HARLOW, H. J., AND G. E. MENKENS, JR. 1986. A com­parison of hibernation in the black-tailed prairie dog, white-tailed prairie dog, and Wyoming ground squir­rel. Canadian Journal of Zoology 64:793-796.

HAYES, S. R. 1976. Daily activity and body tempera­ture of the southern woodchuck, Marmota monax monax, in northwestern Arkansas. Journal of Mam­malogy 57:291-299.

HELDMAIER, G., S. ORTMANN, AND G. KORTNER. 1993. Energy requirements of hibernating alpine marmots. Pp. 175-183 in Life in the cold: ecological, physi­ological, and molecular mechanisms (C. Carey, G. L. Fiorant, B. A. Wunder, and B. Horwitz, eds.). Westview Press, Boulder, Colorado.

LUECKE, R. H., AND F. E. SOUTH. 1972. A possible model for thermoregulation during deep hibernation. Pp. 577-605 in Hibernation and hypothermia: per­spectives and challenges (F. E. South, J. P. Hannon, J. R. Willis, E. T. Pengelley, and N. R. Alpert, eds.). American Elsevier Publishing Company, New York.

LYMAN, C. P. 1958. Oxygen consumption, body tem­perature and heart rate of woodchucks entering hi­bernation. American Journal of Physiology 194:83-91.

LYMAN, C. P., J. S. WILLIS, A. MALAN, AND L. C. H. WANG. 1982. Hibernation and torpor in mammals and birds. Academic Press, New York.

310 JOURNAL OF MAMMALOGY Vol. 84, No.1

MICHENER, G. R. 1977. Effect of climatic conditions on the annual activity and hibernation cycle of Rich­ardson's ground squirrels and Columbian ground squirrels. Canadian Journal of Zoology 55:693-703.

MICHENER, G. R. 1992. Sexual differences in over-win­ter torpor patterns of Richardson's ground squirrels in natural hibernacula. Oecologia 89:397-406.

NAGY, K. A. 1989. Doubly-labeled water studies of vertebrate physiological ecology. Pp. 270-287 in Stable isotopes in ecological research (P. W. Rundel, J. R. Ehleringer, and K. A. Nagy, eds.). Springer­Verlag, New York.

PENGELLEY, E. T. 1967. The relation of external con­ditions to the onset and termination of hibernation and estivation. Pp.I-29 in Mammalian hibernation III (K. C. Fisher, A. R. Dawe, C. P. Lyman, E. Schonbaum, and F. E. South, eds.). Oliver and Boyd, London, United Kingdom.

PiTrENDRIGH, C. S. 1960. Circadian rhythms and the circadian organization of living systems. Cold Spring Harbor Symposium on Quantitative Biology 25:159-184.

PIVORUN, E. B. 1976. A biotelemetry study of the ther­moregulatory patterns of Tamias striatus and Euta­mias minimus during hibernation. Comparative Bio­chemistry and Physiology, A. Comparative Physi­ology 53:265-271.

SNYDER, R. L., AND J. J. CHRISTIAN. 1960. Reproduc­tive cycle and litter size of the woodchuck. Ecology 41:647-656.

SNYDER, R. L., D. E. DAVIS, AND J. J. CHRISTIAN. 1961. Seasonal changes in the weights of woodchucks. Journal of Mammalogy 42:297-312.

WANG, L. C. H. 1972. Circadian body temperature of Richardson's ground squirrel under field and labo­ratory conditions: a comparative radiotelemetric study. Comparative Biochemistry and Physiology, A. Comparative Physiology 43:503-510.

WANG, L. C. H. 1973. Radiotelemetric study of hiber­nation under natural and laboratory conditions. American Journal of Physiology 224:673-677.

WANG, L. C. H. 1978. Energetic and field aspects of mammalian torpor: the Richardson's ground squirrel. Pp. 109-145 in Strategies in the cold: natural tor­pidity and thermogenesis (L. C. H. Wang and J. W. Hudson, eds.). Academic Press, New York.

WANG, L. C. H., AND J. W. HUDSON. 1971. Temperature regulation in normothermic and hibernating eastern chipmunk, Tamias striatus. Comparative Biochem­istry and Physiology, A. Comparative Physiology 38:59-90.

ZERVANOS, S. M. 1969. Effects of photoperiod on the daily activity of the Allegheny wood rat (Neotoma floridana). Mammalia 33:509-515.

ZERVANOS, S. M., AND R. E. HENSHAW. 1970. Graded effect of acclimation temperature on thermogenesis in winter-captured little brown bats. Pennsylvania Academy of Science 44:207-211.

ZERVANOS, S. M., AND C. M. SALSBURY. 2001. Telem­etry system for the continuous monitoring of tem­perature, location and activity of free-ranging ani­mals. Bulletin of the Ecological Society of America 82:189-191.

Submitted 11 June 2001. Accepted 1 April 2002.

Associate Editor was Thomas E. Tomasi.